Improved Estimation of Soil Organic Carbon Storage Uncertainty Using First-Order Taylor Series Approximation

Abstract

Assessment of soil organic C (SOC) stocks is important for monitoring the effect of land use change in the C cycle and for formulation of C sequestration strategies in the context of global climate change. Discrepancies among the recent global SOC estimates by different researchers underscore the importance of precise estimation of the uncertainty associated with the SOC stocks. A method was recently proposed to estimate the SOC storage uncertainty using the Taylor series of approximations. Here we show that the accuracy of SOC storage uncertainty can be improved by incorporating the covariance among the input variables. Measurement of input variables from independent samples or use of an incomplete model leads to either over or underestimation of the SOC storage uncertainty. The application of the method to an experimental data set indicated that ignoring covariance would lead to a substantial overestimate of the uncertainty.